Novel PromptLock ransomware developed by NYU researchers

CyberScoop reports that the Golang-based PromptLock ransomware, the first-ever artificial intelligence-powered ransomware discovered by ESET analysts, was confirmed to have been developed by researchers at New York University's Tandon School of Engineering in a bid to show the dangers of harnessing large language models in illicit payloads.

Only binary-embedded natural language prompts are needed by PromptLock to produce polymorphic variants based on targeted environments, where it could enable completely automated reconnaissance, payload creation, and extortion, according to NYU researchers.

"To our knowledge, we are the first work to demonstrate a fully closed-loop LLM orchestrated ransomware attack with targeted payloads and personalized extortion tactics, along with a comprehensive behavioral evaluation to promote future defenses," said NYU researchers, echoing previous findings from ESET.

Such a development comes after Anthropic's Claude LLM was found to have been exploited to enable widespread data theft and "psychologically targeted" extortion.